Replying to @deepfates
0
0
0
Replying to @fidjissimo
0
0
1
Replying to @Aryvyo
@Aryvyo For comparison, yesterday I built a full-stack unified video gen app for Veo and Sora from scratch, no framework. All api features, job queue, database. Total: 163k tokens (in + out), Cost: about $1.50.
github.com/shermify/video…
2
0
3
Replying to @colin_fraser
Try doing this lol. It’s a good exercise to learn how to prompt the raw video model.

In a black, horizonless world a man is sitting facing the camera with a table in front of him. A glass cup filled with water is on the table in front of him. he calmly lifts his right hand to just below shod and the water in the glass cup becomes weightless. The water in the glass cup gently leaves the cup and rises into the air, forming an amorphous sphere, unaffected by gravity, levitating ~0.5m above the table;
1
0
0
Replying to @Teknium
0
0
1
Replying to @fluopoika
0
0
3
Replying to @amasad
1
0
0
Replying to @basedjensen
@basedjensen ✅ Allow OpenAI to train on your data
0
0
13
Replying to @basedjensen
0
0
0
Replying to @olivercingl
0
0
1
Replying to @tekbog
@tekbog Found it
0
0
2
Replying to @ryanwinchester
0
0
5
Replying to @VictorTaelin
@VictorTaelin I just got in a long argument where it admitted the top was open, filled w water but you still can’t drink from it. Started getting aggressive and drew this diagram lmao
2
1
85
Replying to @colin_fraser
@colin_fraser Just put the logarithmic scale on the x-axis, duh
0
0
1
Replying to @tekbog
0
0
1
Replying to @RichardHanania
0
0
2